A METHOD OF DATA SOURCE GENERATION FOR TESTING DATA MINING ALGORITHMS

Because of the problem of secrecy, diversity of data and so on, data resources for testing data mining algorithms always persecute the research of data mining. Therefore, a method of data source generation based on a cloud model is presented in this paper. The generation algorithm for quantitative data is given and the generation of categorical data is also discussed. Because of the randomness and fuzziness in the cloud model, the generated data contain potential knowledge except known knowledge, and the data distribution and relationship between attributes also have random and fuzzy facts. Researchers can use this model to generate diverse type and diverse quantitative data they want and then test the data mining algorithms.